JMIR Publications

ABSTRACT

Background:

Unipolar major depressive disorder (MDD) and bipolar disorder are two major mood disorders. The two disorders have different treatment strategies and prognoses. However, bipolar disorder may begin with depression and could be diagnosed as MDD at the initial stage which may contribute to treatment failure later on. Previous studies indicated that a significant proportion of patients who were diagnosed with MDD will develop bipolar disorder over time. This kind of hidden bipolar disorder may contribute to the treatment resistance observed in MDD patients.

Objective:

In this population-based study, our aim is to investigate the rate and risk factors for a diagnostic change from unipolar MDD to bipolar disorder during a 10-year follow up. Furthermore, a risk stratification model was also developed for MDD to bipolar conversion.

Methods:

We conducted a retrospective cohort study involving patients who were newly diagnosed with major depressive disorder between January 1, 2000 and December 31, 2004 by using Taiwan National Health Insurance Research Database (NHIRD). All depression patients were observed until (1) diagnosed with a bipolar disorder by a psychiatrist, (2) death, or (3) December 31, 2013. All depression patients were divided into two groups, converted group and non-converted group, based on whether the patients were diagnosed with bipolar disorder during the follow-up period. Six groups of variables within the first six months of enrollment, including personal characteristics, physical comorbidities, psychiatric comorbidities, health care usage behaviors, severity of disorder, and use of psychotropics, were extracted, and were put into CART analysis to generate the risk stratification model for MDD to bipolar conversion.

Results:

There are 2820 MDD patients enrolled in our study. During follow-up period, 536 patients were diagnosed with bipolar disorder (conversion rate = 19.0%). The CART method identified 5 variables (kinds of antipsychotics use within first six months of enrollment, kinds of antidepressant use within first six months of enrollment, total psychiatric outpatient visits, kinds of benzodiazepine use within one visit, and use of mood stabilizer) as significant predictors of risk of bipolar conversion. This CART risk tree was able to stratify patients into high, medium, and low risk for bipolar conversion. In high the risk group, 61.5–100% of depressive patients eventually developed bipolar disorder. Relatively, only 6.4–14.3% of depressive patients developed bipolar disorder in the low risk group.

Conclusions:

The CART method is robust to identify five variables as significant predictors of bipolar conversion. In a simple 2-to 4-step process, these variables permit identification of patients with low, intermediate, or high risk for bipolar conversion. The developed model can be applied to routine clinical practice for early diagnosis of bipolar disorder.